optimized time fetch during flush
This commit is contained in:
@@ -116,14 +116,14 @@ void tcpRecvWatchdog::messageArrivalNotify ( const epicsTime & currentTime )
|
||||
// The send watchdog will be responsible for detecting
|
||||
// dead connections in this case.
|
||||
//
|
||||
void tcpRecvWatchdog::sendBacklogProgressNotify ()
|
||||
void tcpRecvWatchdog::sendBacklogProgressNotify ( const epicsTime & currentTime )
|
||||
{
|
||||
// We dont set "beaconAnomaly" to be false here because, after we see a
|
||||
// beacon anomaly (which could be transiently detecting a reboot) we will
|
||||
// not trust the beacon as an indicator of a healthy server until we
|
||||
// receive at least one message from the server.
|
||||
this->responsePending = false;
|
||||
this->timer.start ( *this, this->period );
|
||||
this->timer.start ( *this, currentTime + this->period );
|
||||
debugPrintf ( ("saw heavy send backlog - reseting TCP recv watchdog\n") );
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user